Complete Rewrite of ISO 10993-1 Creates Industry Buzz

Biological evaluation for all medical devices is conducted using guidance provided in ISO 10993-1, the overarching standard for this process. It is the document that gives the general framework for risk assessment and assists in the selection of appropriate biological endpoints for evaluation. It also describes how available information should be applied to determine the need for further testing. ISO 10993-1 is an important reference document for all medical devices and, therefore, it is no surprise that upcoming revisions to this standard have created quite a buzz in the industry.

Medical device manufacturers may have some big questions regarding this revision, such as how it will impact their current biological evaluation documents, if there are additional aspects requiring evaluation not currently included in their assessments, and if they need to redo any testing or conduct gap assessments to their biological evaluation that was conducted per ISO 10993-1:2018. Let’s discuss some changes and address these points in a bit more detail.­

A complete rewrite

First, it’s important to highlight that the proposed draft for the next revision of ISO 10993-1 represents a complete overhaul. The transition from the 2009 version to the current 2018 included many updates and a notable shift toward a more scientific assessment process. However, this draft is essentially a brand-new document. There are of course many aspects that have been transferred from the current standard to the draft, but (spoiler alert) not even Table A.1 is immune to these sweeping changes.

The latest draft of the normative text incorporates far more detail on the approach to biological evaluation, with extensive portions of current Annex B transferred to the main body of the standard. The draft also takes a strong position on further aligning the standard to general risk management standard ISO 14971.

Additionally, the document has been restructured to make it easier to follow the recommended flow for biological evaluations: beginning with a plan (the main focus of the document), performing the testing, and finally ending with a summary report to tie everything together. One key point emphasized in the draft is the need to periodically review and update the overall summary report on a medical device throughout its lifecycle, using data gathered from post-market feedback and other sources. This is an important aspect of the biological safety of devices and will allow better oversight and mitigation measures where new or changed biological risks are identified.

As outlined in the current standard, the first step in the process is the generation of a biological evaluation plan, a written document that reviews all available data on the device and discusses the strategy that will be used to address relevant biological endpoints (or ‘biological effects’ per the new draft). This could mean addressing them with available datasets or indicating the need to generate additional ones. Of note, the new document emphasizes that if additional animal tests are recommended, then a justification shall accompany this request. While this is a positive step in the industry’s shift away from pre-clinical animal studies to in vitro testing alternatives, it is known that some regulatory regions do not yet acknowledge certain in vitro methods. As a result, conflicting justifications could occur and possibly even scenarios where manufacturers find themselves in the undesirable position of needing to choose more than one test method to address an endpoint. Clearly, not an ideal use of time or resources.

Other aspects of the restructure have enhanced the step-by-step course of the biological evaluation process. Although probably utilized already by many users of this document, the new draft stresses that the description and intended use should be covered first, to help with the understanding of the device and its associated risks. After that, categorization should be defined to select the appropriate biological effects and datasets required to address the biological risk estimation and evaluation. The draft also highlights that when the evaluation is targeted to address the impact of a specific change, the relevant biological risks should then be defined based on the contact type and duration of that specific component under review. This helps to define the appropriate risks that should be addressed for the change.

Cumulative use calculation

Another update to the standard is a clearer description of how to calculate cumulative use. Discrepancies between how some manufacturers count for cumulative contact and how it is done by various regulatory bodies have led to many difficult conversations during regulatory submissions. In a number of cases, this discordance has resulted in submission comments and additional work, sometimes even additional testing, on the part of the manufacturer. The draft does elaborate on this matter and states that when calculating the total exposure period, the period (in days) between the first and last use of a medical device should be taken into consideration. Some straightforward examples are also provided, such as wound dressings being swapped for new ones after x many days of use, where total exposure is then the complete duration of the wound site treatment with the dressing. However, further information is needed to cover cumulative contact for devices that are only used for short periods of time over many days (e.g., for 10 minutes, 2 times a week over the course of 6 months). In this case, accounting for the duration between the first day and last day of treatment as the total exposure may be an over-exaggeration.

In addition to this clarification, one large item that has been brought into the document is the term ‘(reasonably) foreseeable misuse,’ whereas before the evaluation was based on ‘intended use’ only. This new term encompasses the potential expanded use beyond the one defined by the manufacturer. One clear example of this is the extended use duration based on, for instance, clinical feedback from similar devices, resulting in a longer duration of exposure than originally intended (e.g., certain breathing assistance devices or wound products). When data is there to indicate this 'off-label’ use, it would be important to consider, especially in relation to appropriate patient safety evaluation and from a toxicological perspective. However, this new definition in the standard is not well-defined and may lead to misinterpretations of what is ‘foreseeable misuse’ and what situations may be considered so improbable that they should not be considered.

The (in)famous Table A.1

A change that will be particularly noticeable to many is that Table A.1 has ceased to exist in its current form. Instead, it is broken up into several separate tables that cover the relevant endpoints for evaluation under each specific categorization. From a flow perspective, this is an improvement to the document as the risks to be considered and discussed are more readily available to the user. The days of using a ruler to locate relevant lines on an oversized table will soon become a distant memory.

This shift to breaking up Table A.1 may also be seen as the next nudge in a move away from the still hauntingly old-school approach that everything listed must be tested. The draft definitively highlights the use of a risk-based approach and the ‘weight-of-evidence’ to draw conclusions on biological safety.

Given the chance, however, I would take this even a step further and remove the tables entirely, moving the whole evaluation into a flowchart that handles contact type and duration in a step-by-step approach. In this approach, each biological endpoint would be discussed based on the specific device, its characteristics, and its intended use. This would allow the assessor to define whether an endpoint is relevant and then choose the right approach to address it, an update that would make the biological evaluation an in-depth story about the device, its risks, and the appropriate mitigation steps to be taken.

Categorizations

For categorizations pertaining to contact type and duration, the draft has removed the term ‘transitory contact’. It now only states that it may be possible for the manufacturer to prepare a written justification that there is no significant biological interaction with the tissues. This aligns with the approach from the current FDA guidance document, which states that if the device has very brief contact, then an assessment should be conducted to determine if testing is needed.

Along these lines, a new section about ‘insignificant risk skin contact’ has been added to the draft that highlights that devices that contact skin during normal handling and are made of materials that are in common use for consumer products, with similar nature of contact, are considered to have no significant risk. For these devices, a simple written justification in the biological evaluation report (BER) will be sufficient. This is taking the current section 5.2.2 a bit further and really honing in on what we’ve been saying all these years: when we touch things with our hands frequently and in everyday situations, and there is no indication of reactivity, it has been proven not to be concerning. I always refer to my mouse and keyboard being in permanent contact with my fingers, or for some people their phones; we have a wealth of evidence of low risk for these types of devices.

Another change in this section, the term ‘surface contacting devices’ has been removed. What’s more, compromised surfaces (like wounds) are now considered under ‘tissue contact’ whereas before this was considered as a separate category under ‘breached/compromised surface’.

Genotoxicity

For applicable biological endpoints, added scrutiny has been included for devices, especially as it pertains to genotoxicity effects. For instance, while the current standard states that for ‘limited contact devices’ only the ones used as part of the extracorporeal circuits require additional genotoxicity assessment, the new draft expands this guidance to all circulating blood-contacting devices. As it reads, this would mean that, for example, all catheters (even if used for one to two hours) would now require this assessment.

Considering that commonly used materials, such as Pebax, are typically used in these devices and are not known to have a genotoxic potential, this may be a bit extreme. If the final guidance does indeed shift to all blood-contacting devices, a better understanding of the reasoning behind the additional genotoxicity requirement will be critical.

We have seen the scrutiny in justifying out of doing some endpoint testing that seems to be irrelevant for some devices, or for new methods to be validated for testing. The same rigor should also be applied to the inclusion of new endpoint requirements. Of course, if available data indicates a concern for genotoxicity that can actually impact the patient, it should be included. That is the main goal of a biological evaluation and what we do — demonstrate that the device is not going to be harmful when used to treat those in need.

Unique device considerations

Despite all the changes that are happening to the document, what I think is important to highlight is that the draft clearly says that due to the diversity of medical devices, it is recognized that testing will not be necessary or practicable for all harms identified for a given medical device. It is indispensable for evaluations that each medical device be considered on its own merits. As such, we need to be diligent in understanding the background information on the device we are evaluating, as well as in our review process for available information to propose a scientifically sound approach to risk assessment.

Discussions in the US working group

As mentioned, this is an umbrella document and affects the entire medical device industry. Due to this wide applicability, many are eagerly participating in the revision process. To give an idea of the impact, more than 1,000 comments were received from various US participants. These comments have been systematically reviewed by the working group to establish consensus in the proposal and will be delivered to the ISO group, where no doubt discussions will continue. For instance, there was a long conversation around risk estimation, brought in by the stronger alignment with ISO 14971, and establishing a standardized approach to addressing it. Especially considering that this concept is now moved to the normative text, additional guidance on this topic would be beneficial before it is rolled out as a requirement.

Overall, based on the discussions coming out of the US side, considerable progress has been made with the document and many of the concepts are on the right track. That being said, the document still needs further revision and clarification in a number of areas to move on to the next level in the standards committee.